Quantitative Susceptibility Mapping (QSM) is an MRI-based technique that sensitively measures in-vivo iron deposition via relaxation and magnetic susceptibility of brain tissue. Iron is essential for brain homeostasis, including oxidative metabolism, formation and maintenance of neural networks, and myelin synthesis. While increased levels of iron deposition occur during normal aging, high levels may have detrimental effects. Previous work has linked excessive brain iron accumulation to oxidative stress, beta-amyloid and tau toxicity, neurodegeneration, and cognitive dysfunction, particularly memory loss. Physical activity, on the other hand, correlates with higher synaptic integrity and memory performance, even in the presence of neuropathology. To date, it is unknown how physical activity may affect iron deposition-related cognition changes. We examined the moderating role of physical activity on the relationship between QSM hippocampal iron deposition and verbal memory in typically aging adults.

62 cognitively unimpaired older adults from the UCSF Memory and Aging Center (age mean(SD) = 78.34(7.28) years; 56% women; education mean(SD) = 17.94(1.72) years; 85% non-Hispanic White) completed neuropsychological testing and brain MRI during annual research visits, followed by Fitbit™ physical activity monitoring for 30 days. Average total daily steps were aggregated. Participants completed 3T Prisma neuroimaging with QSM, and regional iron deposition levels were quantified. All subjects also underwent diffusion tensor imaging (fractional anisotropy). Verbal memory was assessed via long delay free recall scores from the California Verbal Learning Test II (CVLT-II). Linear regression examined verbal memory as a function of hippocampal QSM (bilateral), physical activity, and their interaction. Models covaried for age, sex, and education. Additional models separately examined left and right hippocampal QSM, as well as subcortical QSM to determine lateralization and specificity of verbal memory effects to hippocampal iron deposition, respectively.

Univariably, higher bilateral hippocampal QSM correlated with worse verbal memory (r= 0.35; p= 0.015). Adjusting for demographics, physical activity moderated the relationship between bilateral hippocampal QSM and verbal memory (ß= 0.41, p= 0.011), such that at higher levels of physical activity, the negative relationship between hippocampal QSM and verbal memory was significantly attenuated. Results persisted when adjusting for DTI integrity of the uncinate fasciculus and fornix white matter tracts. Lateralization models were both significant, suggesting that results were not dominantly driven by either left (ß= 0.34, p= 0.048), or right (ß=0.31, p= 0.035) hippocampal QSM. In contrast, subcortical QSM did not correlate with memory performance (r= 0.13, p > 0.05) or interact with physical activity on verbal memory outcomes (p > 0.05).

Physical activity significantly moderated the negative relationship between hippocampal QSM and verbal memory performance. Higher exercise engagement may buffer the adverse effect of hippocampal iron deposition on memory, potentially through its role in maintenance of myelin and synaptic integrity and/or protecting against other neurotoxic events (e.g., oxidative stress, neuronal cell death). Our results support that physical activity continues to be a modifiable risk factor that may offer a protective role in neurobiological pathways of memory and cognitive decline.